158 



cases there are four. There are 15 similar specimens in which the new 

 rays are more than 5 mm long; ten of these have four new rays and 

 five have five. In all these 44 specimens, the old or parent ray is from 

 18 to 41 mm in length, with the diameter ranging from 15 to 20 per cent 

 of the length. An unusually large and line specimen, in this "comet" 

 stage of development, has the parent ray 93 mm long and the younger 

 rays are 62, 62, 54, 60, 56 mm; the rays are each about 11 mm in 

 diameter. 



The remaining 45 specimens are strikingly irregular in form. The 

 largest has five rays 130, 120, 68, 64 and 63 mm long respectively, and 

 13 mm in diameter. Eighteen others have five rays while twenty-four 

 have six and one has seven. In all of these specimens, no two of the 

 arms are of exactly the same length ; often two or three rays will be 

 approximately equal and the others very short, just beginning to re- 

 generate. One of the interesting cases is a large specimen with two 

 arms about 105 mm long, a third 80 mm, a fourth 58 mm, a fifth 39 mm, 

 and a sixth which was severed 13 mm from the mouth but has a re- 

 generating tip 4 mm long. 



On the reefs of Jamaica, Linckia guildingii occurs on the under 

 side of fragments of coral rock, which are well exposed to the surf. 

 Here, protected from the surf but surrounded by the abundance of or- 

 ganisms which swarm in such well-aerated places, it leads a very slug- 

 gish life. In very shallow water only the smallest specimens are found; 

 the larger specimens occur further out. The full-grown specimens seem 

 to be confined to relatively deep Avater, for I have myself only once col- 

 lected a full-grown adult. My other specimens have been brought to 

 me by native fishermen. 



Taking into consideration all the facts now known I think we are 

 driven to the conclusions which I summarize below. Some of them are 

 not capable of positive proof but the weight of evidence justifies our ac- 

 cepting them. 



1) Autotomy in Linckia is not merely a normal process originated 

 by internal stimuli but in L. guildingii at least, is an asexual method 

 of rej)roduction of prime importance. 



2) Young Linckias have five or six (sometimes possibly seven) rays 

 of approximately equal length; the number is not subsequently either 

 increased or decreased. 



3) When the rays are about 15 mm long, autotomy begins to occur 

 more or less irregularly; usually only one ray is severed but frequently 

 two, rarely three or possibly four. 



4) Growth continues after autotomy, uninterruptedly, but most 

 rapidly in the regenerating rays. 



